Clinical and Pharmacological Considerations for Concomitant Administration of Posaconazole and Isavuconazole with Letermovir

ABSTRACT

We sought in this case-control retrospective study to compare posaconazole and isavuconazole (PCZ and IVC, respectively) plasma trough concentration (C_trough) levels in high-risk allogeneic hematopoietic cell transplant (HCT) recipients who received letermovir (LET) or not. PCZ/IVC C_trough levels were not found to be significantly different between cases and controls, as they were 1.31 mg/liter (median) (interquartile range [IQR], 0.90) versus 1.36 mg/liter (IQR, 1.16) (P = 0.31) and 3.20 mg/liter (IQR, 2.40) versus 2.35 mg/liter (IQR, 1.50) (P = 0.17), respectively. In conclusion, we observed PCZ/IVC C_trough levels within the expected range and no significant effect of LET coadministration.

TEXT

Broad-spectrum azoles, including voriconazole (VCZ), posaconazole (PCZ), and isavuconazole (IVC), are commonly used as prophylaxis or treatment in high-risk allogeneic hematopoietic cell transplant (HCT) recipients (1, 2). Azoles are well-known inhibitors but also substrates of a wide range of cytochrome P (CYP) enzymes, with a high potential for being victims of drug-drug interactions (DDI) (3–5). Letermovir (LET) is currently used as primary prophylaxis for the first 100 days after an allogeneic HCT in cytomegalovirus (CMV) serology-positive recipients (6). LET is a moderate inhibitor of CYP3A4/5, CYP2C8, organic anion transporting polypeptide 1B1 and 3 (OATP1B1/3), and P-glycoprotein (P-gp) and may induce CYP2C19 and CYP2C9 (7, 8). LET has been reported to decrease the VCZ area under the curve (AUC) by almost 50% in healthy individuals and 33% in allogeneic HCT recipients (9–11). No significant effect of LET on PCZ pharmacokinetics has been reported in healthy volunteers, but there are no available data in HCT recipients (8). There are no available data on the effect of LET on IVC.

Considering the aforementioned CYP effects and interactions, we hypothesized that coadministration of LET may have no significant effect on plasma trough concentration (C_trough) levels of PCZ and IVC. To test this hypothesis, we performed a case-control study to compare PCZ and IVC C_trough distribution between patients with and without LET coadministration.

The study was approved by the local ethics committee. Participants were identified through the institutional HCT and pharmacy databases. Case patients were included in this study if they were adult (≥18-year-old) allogeneic HCT recipients, who received LET with PCZ (intravenous [i.v.] and/or delayed-release tablets [DRTs]) or IVC (i.v. and/or orally [p.o.]) as prophylaxis or treatment with at least 1 PCZ/IVC C_trough level available at least after 5 days of azole administration. Institutional practices can be found in the supplemental material. Considering that time to steady state is approximately 10 days for LET, probably due to saturation of OATP-mediated uptake in the liver, only PCZ/IVC C_trough levels obtained after 10 consecutive days of LET administration were included for the analyses (7, 9). Patients were excluded if they received PCZ suspension or any azole for <7 days and had no available azole C_trough level after day 5 of azole administration. VCZ was not considered for the study, as it is less used in our center. Controls were consecutive adult allogeneic HCT recipients who received PCZ (i.v. and/or DRT) or IVC (i.v. and/or p.o.) as prophylaxis or treatment for a minimum of 7 consecutive days without concomitant administration of LET. Cases were matched 1:1 to controls for the following variables: (i) antifungal agent (PCZ or IVC administration), (ii) azole dose and mode of administration, (iii) age (less than and ≥60 years), (iv) gender, and (v) time since HCT (less than and ≥15 days post-HCT). The selection of 15 days post-HCT as a matching criterion was based on the presumed potential pharmacokinetic changes over time after HCT, particularly those in the early period after HCT and associated with the presence of conditioning-induced gastrointestinal mucositis and neutropenia, which were considered to occur, on average, by post-HCT day 15 (12). PCZ/IVC C_trough concentrations were performed in the context of routine clinical management of allogeneic HCT recipients. PCZ therapeutic drug monitoring (TDM) was performed in plasma samples with high-performance liquid chromatography-mass spectrometry (HPLC-MS) by Chromsystems Instruments & Chemicals (Gräfelfing, Germany). IVC TDM was performed in plasma samples with liquid chromatography with tandem mass spectrometry (LC-MS/MS; Recipe 200 kit system; Germany). Categorical and continuous variables were compared with the Fisher’s exact test and a two-tailed Student's t test or Mann-Whitney test when appropriate, respectively. Intraindividual and interindividual variability of PCZ/IVC levels were assessed by calculating the median of the coefficient of variation (CV) of all PCZ/IVC levels measured in a single patient and the CV of the average PCZ/IVC levels of each patient, respectively. A 2-sided P value of <0.05 was considered statistically significant for all tests. Statistical analysis was performed using Stata 16.0 (StataCorp., College Station, TX).

We identified 16 and 8 patients, who received LET concomitantly with PCZ and IVC, respectively, with available drug TDM. Two patients who received PCZ and LET were excluded because LET and PCZ were deemed to not have reached steady state when PCZ C_trough levels were obtained. Two patients were excluded from the IVC group because of the lack of matched controls. The characteristics of cases and controls are detailed in Table 1. Cases appeared to be well matched to controls for their major baseline characteristics. Thirty and 29 PCZ C_trough levels were identified for cases and controls, respectively. The PCZ C_trough was not significantly different in cases (median, 1.31 mg/liter; interquartile range [IQR], 0.90) compared to controls (median, 1.36 mg/liter; IQR, 1.16; P = 0.31) (Fig. 1A). Twelve IVC C_trough levels were identified in each group. The IVC C_trough was not significantly different in cases (median, 3.20 mg/liter; IQR, 2.40) compared to controls (median, 2.35 mg/liter; IQR, 1.50; P = 0.17) (Fig. 1B). In total, 16.66% and 20.69% (P = 0.70) of the PCZ C_trough levels were <0.7 mg/liter for cases and controls, respectively. All IVC C_trough levels were >1.0 mg/liter for both cases and controls. Intra- and interindividual variability of PCZ C_trough levels were similar between cases and controls at 43.05% versus 48.34% and 46.54% versus 49.51%, respectively. Similar observations were made for IVC C_trough intra- and interindividual variability between cases and controls at 32.94% versus 30.00% and 46.13% versus 67.72%, respectively.

TABLE 1.

Case and control characteristics for posaconazole and isavuconazole

Characteristicc	Posaconazole			Isavuconazole
	Cases (n = 14)	Controls (n = 14)	P value	Cases (n = 6)	Controls (n = 6)	P value
Demographics
Age (mean [range] [yrs])	55.9 (22 to 70)	46.2 (17 to 72)	0.13	60.2 (18 to 78)	55 (37 to 75)	0.65
Age <60 yrs (no. [%])	6 (42.9)	10 (71.4)	0.13	4 (33.3)	3 (50.0)	0.56
Gender, female (no. [%])	9 (64.3)	9 (64.3)	1.00	2 (33.3)	1 (16.7)	0.51
HCT characteristics (no. [%])
Conditioning, reduced intensity	14 (100)	13 (92.9)	0.31	5 (83.3)	6 (100)	0.30
HCT donor	0.71	0.77
HLA matched, related	2 (14.3)	3 (21.4)	2 (33.3)	2 (33.3)
HLA matched, unrelated	9 (64.3)	6 (42.9)	3 (50)	2 (33.3)
Haploidentical	3 (21.4)	5 (35.7)	1 (16.7)	2 (33.3)
HCT source (no. [%])	0.40	0.12
Bone marrow	3 (21.4)	5 (35.7)	0	2 (33.3)
Peripheral blood	11 (78.6)	9 (64.3)	6 (100)	4 (66.7)
Acute GvHD grade ≥2a	10 (71.4)	6 (42.9)	0.13	2 (33.3)	1 (16.7)	0.51
CMV serology status (no. [%])	0.68	0.25
Donor negative/recipient negative	6 (42.9)	4 (28.6)	0	0
Donor negative/recipient positive	3 (21.4)	3 (21.4)	0	0
Donor positive/recipient negative	0	1 (7.1)	2 (33.3)	4 (66.7)
Donor positive/recipient positive	5 (35.7)	6 (42.9)	4 (66.7)	2 (33.3)
Azole administration
Indication for azole (no. [%])	0.62	1.00
Prophylaxis	11 (78.6)	12 (85.7)	0	0
Treatment of possible/probable IFI	3 (21.4)	2 (14.3)	6 (100)	6 (100)
Loading dose, yes (no. [%])	13 (92.9)	11 (78.6)	0.30	6 (100)	6 (100)	1.00
Azole mode of administration, i.v. (no. [%])	4 (28.6)	4 (28.6)	1.00	3 (50)	3 (50)	1.00
Posaconazole, 300 mg once daily (no. [%])	12 (85.7)	12 (85.7)	1.00	NA	NA
Isavuconazole, 200 mg once daily (no. [%])	NA	NA	6 (100)	6 (100)	1.00
Azole initiation day post-HCT (median [IQR])b	20.5 (−6 to 175)	9.5 (−1 to 247)	0.76	9 (−19 to 29)	−14.5 (−148 to 16)	0.19
1st TDM ≥15 days post-HCT (no. [%])	11 (78.6)	7 (50.0)	0.12	4 (66.7)	3 (50)	0.56

^aInformation on GvHD was recorded at study inclusion.

^bNegative values suggest that azole initiation was before HCT.

^cHCT, hematopoietic cell transplant; HLA, human leukocyte antigen; GvHD, graft versus host disease; CMV, cytomegalovirus; IQR, interquartile range; IFI, invasive fungal infection; i.v., intravenous; TDM, therapeutic drug monitoring; NA, not available.

FIG 1.

(A) Distribution of posaconazole plasma trough concentration (C_trough) levels in 14 cases and 14 controls. There was no statistically significant difference between the median posaconazole C_trough between cases and controls (P = 0.31). (B) Distribution of isavuconazole C_trough in 6 cases and 6 controls. There was no statistically significant difference between the median isavuconazole C_trough between cases and controls (P = 0.17). Boxes represent the median and 25th and 75th percentiles; whiskers represent the range of maximum and minimum values within the interquartile range. Outliers are shown.

To our knowledge, this is the first study to assess the potential effect of LET on PCZ/IVC C_trough in allogeneic HCT recipients. It is also the first to study the effect of LET on IVC in vivo. We observed PCZ and IVC C_trough levels within the expected range and no significant effect of LET coadministration on PCZ and IVC concentrations. Compared to controls, there were no significant differences in PCZ/IVC C_trough levels in patients who received PCZ/IVC concomitantly with LET, as suggested by the preexisting in vitro and in vivo limited data, and they are consistent with the metabolism pathways of these agents (4, 5, 8). As previously reported by our group and others (13–20), our results revealed a high intraindividual variability of plasma concentrations of PCZ and IVC. This variability can be explained by several factors, including (i) pharmacokinetic changes known to occur over time in HCT recipients (19), (ii) mode of administration (i.v. versus p.o.), or (iii) differences in the pharmacogenetic characteristics of patients and DDIs. One strength of this study was to consider azole plasma concentrations once LET and azoles had reached a steady state, limiting intraindividual variability. However, the conclusions of this single-center study should be interpreted with caution in the light of several limitations, including, but not limited to, the (i) small number of included patients; (ii) 1:1 matching between cases and controls, considering the small number of eligible HCT patients in our center; (iii) matching criteria that did not include all treatments with potential interactions with azoles and/or LET; and (iv) potential pharmacogenetic polymorphisms not explored. In conclusion, we report that concomitant administration of LET with PCZ and IVC in allogeneic HCT recipients does not significantly affect the PCZ and IVC plasma concentrations. Further prospective dedicated studies are required to confirm these results.